While an intellectually active and socially integrated lifestyle shows promise for promoting cognitive resilience, the mechanisms underlying any such effects are not well understood. Importantly, the effects of cognitive training on measures of cognitive outcomes have been clearly shown to produce very narrow effects. At the same time, intelligence ? and cognitive health ? manifest as a ?positive manifold? of correlated abilities. Thus, the scientific puzzle to be solved is how in the ecology of ordinary life this positive manifold emerges in a system that appears to be built to respond to experience very narrowly. While conventional models of intelligence have viewed the positive manifold as reflective of a latent construct, recent models of intellectual function suggest that the positive manifold may emerge out of components that reciprocally support and influence one another (?mutualism?). Such a view is also consistent with neural models suggesting that structural and functional brain networks are reorganized (even on fairly short timescales) in response to experience so as to support intelligence. Such a view implies that improvement in one component may enhance the modifiability of a related component. This has not been tested in conventional assessments of transfer that examine the effects of training on measures of function at a single time point. We propose an Ecological Model of Enrichment, in which mutualism plays a key role in synergizing cognitive resilience with aging. We test this model in a series of experiments based on a ?successive-enrichment? paradigm in which we examine improvement in cognitive skills, as well as changes in structural and functional connectivity, as a function of prior training-related improvements on related skills. We instantiate this by examining interactive effects of working memory and language competence.